# AudSemThinker: Enhancing Audio-Language Models through Reasoning over Semantics of Sound

arXiv: https://arxiv.org/abs/2505.14142 · HTML: https://mlsanity.com/paper/2505.14142/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 May 2025.

## Built on by (most cited first, 11 of 11)

- [Audio-Thinker: Guiding Audio Language Model When and How to Think via Reinforcement Learning](https://mlsanity.com/paper/2508.08039/) (2025)
- [The Interspeech 2026 Audio Reasoning Challenge: Evaluating Reasoning Process Quality for Audio Reasoning Models and Agents](https://mlsanity.com/paper/2602.14224/) (2026)
- [Audio Flamingo Sound-CoT Technical Report: Improving Chain-of-Thought Reasoning in Sound Understanding](https://mlsanity.com/paper/2508.11818/) (2025)
- [Omni-CLST: Error-aware Curriculum Learning with guided Selective chain-of-Thought for audio question answering](https://mlsanity.com/paper/2509.12275/) (2025)
- [Encode Once, Decode Never: Reusing Audio LM Internals for Efficient Temporal Localization](https://mlsanity.com/paper/2602.10230/) (2026)
- [Learning When to Think While Listening in Large Audio-Language Models](https://mlsanity.com/paper/2605.27190/) (2026)
- [SightSound-R1: Cross-Modal Reasoning Distillation from Vision to Audio Language Models](https://mlsanity.com/paper/2509.15661/) (2025)
- [HumMusQA: A Human-written Music Understanding QA Benchmark Dataset](https://mlsanity.com/paper/2603.27877/) (2026)
- [Beyond Transcription: Unified Audio Schema for Perception-Aware AudioLLMs](https://mlsanity.com/paper/2604.12506/) (2026)
- [BASS: Benchmarking Audio LMs for Musical Structure and Semantic Reasoning](https://mlsanity.com/paper/2602.04085/) (2026)
- [Empowering Long-form Omni-modal Understanding with Robust Audio Perception](https://mlsanity.com/paper/2607.10299/) (2026)

## Abstract (arXiv)

Audio-language models have shown promising results in various sound understanding tasks, yet they remain limited in their ability to reason over the fine-grained semantics of sound. In this paper, we present AudSemThinker, a model whose reasoning is structured around a framework of auditory semantics inspired by human cognition. To support this, we introduce AudSem, a novel dataset specifically curated for semantic descriptor reasoning in audio-language models. AudSem addresses the persistent challenge of data contamination in zero-shot evaluations by providing a carefully filtered collection of audio samples paired with captions generated through a robust multi-stage pipeline. Our experiments demonstrate that AudSemThinker outperforms state-of-the-art models across multiple training settings, highlighting its strength in semantic audio reasoning. Both AudSemThinker and the AudSem dataset are released publicly.